Azure

Virtual Machines Insights

I recently worked on implementing VM Insights into Virtual Machines. However the latest templates provided by Microsoft were still written in ARM and generally a little complex. I have converted them to Bicep and edited them a little bit.

param location string = resourceGroup().location

param virtualMachineName string
param logAnalyticsNamespaceName string

resource virtualMachine 'Microsoft.Compute/virtualMachines@2023-07-01' existing = {
  name: virtualMachineName
}

resource logAnalyticsWorkspace 'Microsoft.OperationalInsights/workspaces@2022-10-01' existing = {
  name: logAnalyticsNamespaceName
}

resource MSVMI_DCR 'Microsoft.Insights/dataCollectionRules@2022-06-01' = {
  name: 'MSVMI-${virtualMachine.name}'
  location: location
  properties: {
    description: 'Standard data collection rule for VM Insights.'
    dataSources: {
      performanceCounters: [
        {
          name: 'VMInsightsPerfCounters'
          streams: [
            'Microsoft-InsightsMetrics'
          ]
          samplingFrequencyInSeconds: 60
          counterSpecifiers: [
            '\\VmInsights\\DetailedMetrics'
          ]
        }
      ]
      extensions: [
        {
          streams: [
            'Microsoft-ServiceMap'
          ]
          extensionName: 'DependencyAgent'
          name: 'DependencyAgentDataSource'
        }
      ]
    }
    destinations: {
      logAnalytics: [
        {
          workspaceResourceId: logAnalyticsWorkspace.id
          name: 'VMInsightsPerf-Logs-Dest'
        }
      ]
    }
    dataFlows: [
      {
        streams: [
          'Microsoft-InsightsMetrics'
        ]
        destinations: [
          'VMInsightsPerf-Logs-Dest'
        ]
      }
      {
        streams: [
          'Microsoft-ServiceMap'
        ]
        destinations: [
          'VMInsightsPerf-Logs-Dest'
        ]
      }
    ]
  }
}

resource Vm_DaExtension 'Microsoft.Compute/virtualMachines/extensions@2023-07-01' = {
  parent: virtualMachine
  name: 'DependencyAgentWindows'
  location: location
  properties: {
    publisher: 'Microsoft.Azure.Monitoring.DependencyAgent'
    type: 'DependencyAgentWindows'
    typeHandlerVersion: '9.5'
    autoUpgradeMinorVersion: true
    settings: {
      enableAMA: 'true'
    }
  }
}

resource Vm_AmaExtension 'Microsoft.Compute/virtualMachines/extensions@2023-07-01' = {
  parent: virtualMachine
  name: 'AzureMonitorWindowsAgent'
  location: location
  properties: {
    publisher: 'Microsoft.Azure.Monitor'
    type: 'AzureMonitorWindowsAgent'
    typeHandlerVersion: '1.0'
    autoUpgradeMinorVersion: true
  }
}

resource Vm_Microsoft_Insights_VMInsights_Dcr_Association 'Microsoft.Insights/dataCollectionRuleAssociations@2022-06-01' = {
  name: '${virtualMachine.name}-VMInsights-Dcr-Association'
  scope: virtualMachine
  properties: {
    description: 'Association of data collection rule for VM Insights.'
    dataCollectionRuleId: MSVMI_DCR.id
  }
}

Health ❤️ ❤️ ❤️ ❤️ ❤️

How to find Virtual Machine SKUs

I found it a little challenging understanding how to obtain the SKU details required to deploy a Virtual Machine using Bicep

The Microsoft.Compute/virtualMachines resource requires a few properties related to the image/sku of the machine.

imageReference: {
    publisher: '?'
    offer: '?'
    sku: ?
    version: 'latest'
}

To obtain these properties some Powershell is required. The examples given in the offical Microsoft documentation weren’t particularly helpful in my opinion.

In this example I’m going to look for a Microsoft Windows 10 image.

Azure Functions Intro

I’ve been playing around with Azure Functions recently. I’m hoping to do a specific write up later explaining what I’m building and why. For now though I’ve been writing some easy Python code to read data from Table Storage and create messages on Queues.

It’s pretty easy to get started using VSCode with the Azure extension to create, write and run the functions. To emulate the storage I’m using Azurite which is an open source Azure storage emulator. This is needed to emulate the functions storage and also the Table and Queue that I’m using to read and write data to.